Studia Geotechnica et Mechanica, Vol. XXXII, No.

3, 2010

BEHAVIOUR OF FIBRE-REINFORCED AND

STABILIZED CLAYEY SOILS SUBJECTED TO CYCLIC LOADING

G.P. DALL’AQUA
AIC Projetti, Rome, Italy. Formerly of the University of Birmingham,
Birmingham, United Kingdom.

G.S. GHATAORA and U.K. LING

School of Civil Engineering, University of Birmingham, Birmingham B15 2TT, United Kingdom.

Abstract: Much research has been undertaken on the use of fibres to reinforce soils for varying range of
applications such as adobe bricks and walls and pavements, but little is available on the response of
these materials to repeated loading the subgrade soils in road pavement may be subjected to. Thus, with
a view on the application of pavement design, an investigation was undertaken to assess the effect of fi-
bre on kaolinite and laterite stabilised with both cement and lime subjected to repeated loading.
Crimped monofilament of 12 mm long polypropylene fibre with a diameter of 18 microns was
used to reinforce both the soils at concentration of 0.3% stabilized with 4% and 6% of lime and ce-
ment. Results show that kaolinite soils reinforced with 0.3% of fibres together stabilized with 6%
cement under repeated axial load test deform less than 1% after 3,600 load cycles and could be used
in pavement construction. For laterite soils under the worst case scenario conditions of soaking, the
samples of plain soil and those stabilized with cement did not have enough strength and collapse be-
fore loading. However, reinforced and stabilised, particularly were strong enough after soaking to be
used in the upper parts of a pavement.

1. INTRODUCTION

Fibres extracted from vegetation of various types have been used for reinforcing
soils for thousands of years and in some parts of the world fibres are still used in the
construction of bricks for low rise buildings. A number of researchers have considered
the effect of using fibres in clayey soils. WIBISONO [14] investigated erosion of fibre-
reinforced soil. Other have looked into the improvement in tensile strength of fibre-
reinforced soils (GHATAORA et al. [7]) and many have looked into the effect of fibres
on strength gain of a range of soils (MAHER and YO [9], RANJAN et al. [11],
DALL’AQUA [4], DALL’AQUA et al. [5], GHATAORA et al. [6], PARK and TAN [10],
YETIMOGLU and SLABAS [13], and TANG et al. [12]). Most of the researchers have
considered the improvements of soil properties under monotonic loading and in many
cases only fibre is added. WIBISONO [14], DALL’AQUA [4] and GHATAORA et al. [7],
[8] looked into the properties of fibre-reinforced soils with a binder. In general terms,
the improvement of the properties of fibre-reinforced stabilized soils is significant. For
instance, DALL’AQUA [4] shows that there is a marginal increase in the strength of
unstabilized kaolinite. The effect of the fibre on the strength of kaolinite is shown in
4 G.P. DALL’AQUA et al.

figure 1, and comparison of plain kaolinite with cement-stabilized kaolinite is shown

in figure 2. In all the cases, the materials display a very brittle behaviour. The im-
provement in kaolinite, when both fibre and cement are added, is shown in figure 3.
There was greater than twofold increase in strength. In addition to this, we observe
a considerable increase in post-peak strength as a significant strength is retained be-
yond the peak strength. Mixtures such as these have also shown a high resistance to
sheet erosion (WIBISONO [14]).

Fig. 1. Load–deformation characteristics of plain and fibre-reinforced kaolinite (DALL’AQUA [4])

Fig. 2. Load–deformation characteristics of plain and cement-stabilized kaolinite (DALL’AQUA [4])

 Behaviour of fibre-reinforced and stabilized clayey soils 5

Fig. 3. Load–deformation characteristics of plain and fibre-reinforced stabilized kaolinite

(DALL’AQUA [4])

Due to the right content and type of fibre in stabilized soil, the resistance to per-
manent deformation can be achieved in pavement design. GHATAORA et al. [8] inves-
tigated the behaviour of fibre-reinforced kaolinite and laterite soils and showed that
inclusion of fibres increased the strength by about 30% and 95%, respectively, when
they were compacted at optimum moisture content (OMC). In addition to this, the
stabilized and fibre-reinforced soil specimens proved to be more durable when soaked,
compared to both stabilized soil (without fibre) and untreated soil.
DALL’AQUA [4] investigated the effect of inclusion of fibres on unstabilized
clayey soils and soils stabilized with both lime and cement. These fibres ranged in
cross-section from 18 μ to 0.3 mm. The properties of soils with fibre content of up to
0.6% were investigated. He proved that the optimum fibre content approached 0.3%
concentration that gave the best performance. Addition of a larger amount of fibres
can lead to their uneven distribution resulting in non-homogeneous mix and conse-
quently in the reduction in strength. He also showed that 12-mm long fibres were best,
since longer fibres proved difficult to mix.
The research reported herein was aimed at investigating the behaviour of both
kaolinite and, to a lesser extent, laterite (due to its limited availability) soils that were
reinforced with fibres and stabilized with cement. A small study was undertaken with
lime as a stabilizer, but it was not expected to be very effective as its reactivity for
strengthening soils is slow. It was envisaged that the main application of such materi-
als would be in road pavement construction.
The design parameter required in addition to strength is the modulus of the soil. If
the fibre-reinforced stabilized soil is to be used at the road surface, then it should meet
6 G.P. DALL’AQUA et al.

the same criteria as the asphalt surface in the case of thin surfacing layer. Since this
related to strain, then if the latter is within acceptable limits for a given load, the mate-
rial is worthy of consideration for use in road pavement layers. Thus it is inferred that
under the worst conditions the results should comply within the maximum value rec-
ommended by COOPER et al. [3] for asphalt mix design, where the maximum axial
strain measured under repeated axial load should be within 1% after 3,600 load cycles.
In addition to this, permanent deformations should be within the maximum acceptable
value when soil is compacted at maximum dry density and optimum moisture content.

2. MATERIALS AND MIXES

2.1. SOILS

Processed kaolinite clay (referred to as “kaolinte” in the text) from only one batch
was used in this study so that the effect of variability due to soil properties could be
reduced. Thus, the effect of inclusion of fibre and cement could be better understood.
Additionally, plain laterite (a natural deposit) was used to understand how a natural
deposit may behave. Index properties of both the soils are given in table 1. Both soils
are silty clays. However, laterite has higher specific gravity than kaolinite.

Table 1
Properties of kaolinite and laterite
Properties Kaolinite Laterite
Liquid limit (%) 42 45
Plastic limit (%) 24 21
Plasticity index (%) 16 24
Specific gravity 2.64 2.87
Optimum moisture content (%) 23 13
Maximum dry density (kg/m3) 1571 2056

All tests were conducted in accordance with BS1377, 1990 (British Standard In-
stitution 1990a).

2.2. FIBRES

Fibre known as “F23” (F23 was manufacturer’s designation and it was supplied by
Fibrin Ltd. UK) gave the overall best performance in terms of unconfined compressive
strength (UCS) and in particular tensile strength (DALL’AQUA [4]). The properties of the
F23 fibre are given in table 2. Soils reinforced with F23 were subjected to repeated load.
 Behaviour of fibre-reinforced and stabilized clayey soils 7

Table 2
Properties of F23 fibres

Properties Value/Description Property Value/Description

Length (mm) 12 Density (g/cm3) 0.91
Diameter (μm) 18 Tensile strength (MPa) 37
Type Micromonofilament Water absorption Nil
Polymer Polypropylene Softening point (oc) 160
Shape Crimped Surface area (cm2/g) 0.24

2.3. LIME AND CEMENT

Clayey soils can soften due to increase in moisture content. The magnitude of the
reduction in strength depends on the type of clay and both the fraction of clay present
and the amount of water added. Thus any reinforcement that relies on both friction and
cohesion for strength will become less effective as water content is increased. In order
to overcome this drawback, the use of both lime (quick lime) and cement (Portland
cement) as binders was considered.
In order to estimate the amount of the lime required, initial lime consumption test
was undertaken in accordance with the BS1924, 1990 (British Standard Institution,
1990b). The results of the tests give an indication of the maximum amount of the lime
required for soil modification. A greater amount of lime will contribute to strength
gain. In the case of kaolinite, the initial consumption of both cement and lime was
approximately 3.2%. Therefore, 4% and 6% concentrations of binders were used with
fibre reinforcement to ensure that their amount was adequate to produce the stabilizing
effect. The same concentrations were used for laterite soils. An initial lime consump-
tion for the laterite was about 2%.

2.4. MIXES INVESTIGATED

Eight mix compositions were examined during this investigation (table 3). All
specimens for cyclic load tests were prepared by using static compaction technique.
(Fixed mass of soil was compacted to obtain a predetermined volume using a hydrau-
lic ram). The specimens were 90 mm high with the diameter of 100 mm to suit the
Nottingham Asphalt Test (NAT) apparatus (supplied by Cooper Research Technology
Ltd. UK). The kaolinite + cement + fibre and kaolinite + fibre + lime samples were
prepared, sealed with cling film and stored in high humidity environment at 20 °C for
28 days. Subsequently, a set of these samples was immersed in water for 7 days for
durability testing. For laterite soils, the samples were cured in cling film (sealed) and
cured for 7 days in a humid environment followed by 4-day immersing in water. (All
the results obtained from soaking tests are referred to as “soaking” in the text).
8 G.P. DALL’AQUA et al.

Table 3
Mix compositions
1 Kaolinite + fibre 5 Kaolinite + fibre + lime
2 Kaolinite + cement 6 Laterite + cement
3 Kaolinite + lime 7 Laterite + fibre + cement
4 Kaolinite + fibre + cement 8 Laterite + fibre

3. TEST RESULTS

In general, four specimens were prepared for each test and normally it was sufficient
to test three specimens since the repeatability of the results was exceptionally good.

3.1. COMPRESSIVE STRENGTH OF KAOLINITE SUBJECTED TO MONOTONIC LOADING

A comparison of dry density–moisture content relationship for kaolinite, kao-

linite+0.3% fibre, kaolinite+6% cement and kaolinite + fibre + cement is shown in
figure 4. The results reveal that the inclusion of cement gives a marginal reduction in
the density of the mixture; whereas fibres alone do not have any significant effect. The
addition of cement results in a small increase in the optimum moisture content of the
soil. Soils were mixed at a range of moisture content and cured for 28 days before
determination of their unconfined compressive strength (UCS). Typical strengths for
6% cement and 0.3% fibre are shown in figure 5. Maximum UCS occurred at mois-
ture-compacted couple of percentages below the optimum moisture content (deter-
mined from a compaction test shown in figure 4).

kaolinite
kaolinite+0.3% F23
kaolinite+6% cement
kaolinite+6% cement+0.3% F23

Fig. 4. Relationship between dry density and moisture content of kaolinite

 Behaviour of fibre-reinforced and stabilized clayey soils 9

Fig. 5. Unconfined compressive strength–moisture content curves for kaolinite plus 6% cement
and kaolinite plus 0.3% fibre and 6% cement

3.2. BEHAVIOUR OF FIBRE-REINFORCED SOILS SUBJECTED TO REPEATED LOADING

Resistance to deformation of the specimens subjected to repeated axial load was

assessed using the NAT apparatus. This apparatus is designed to measure the defor-
mation properties of asphalt under cyclic loading. Since fibre-reinforced soils were
being considered for upper pavement layers, NAT was considered to be suitable for
examining the behaviour of the materials. As no literature has been found on the study
of plain or stabilized soil using the NAT apparatus, the results of this investigation are
assessed in terms of the results for asphalt mixture. NAT allows us to subject speci-
mens to load cycle which is two seconds long: one-second load period is followed by
one-second rest period. Prior to the actual test, the specimen was subjected to a ten-
minute initial conditioning where an axial stress of 10 kPa is applied. This is carried
out to bed the platens and thus to reduce the bedding-in deformation errors at the be-
ginning of the test. In accordance with this, following the initial conditioning period,
the specimen height varies, depending on the specimens with different amount of fibre
or stabilizer. With the initial conditioning load likely to cause initial deformations, the
test is based on actual strains not taking into consideration the strain induced during
the initial conditioning period. All the specimens were prepared at optimum moisture
content and maximum dry density.
Test conditions implemented were as follows:
Axial test stress, 100 kPa.
Conditioning stress, 10 kPa.
Conditioning period 10 minutes.
10 G.P. DALL’AQUA et al.

Test duration, 10,000 load cycles

Load cycle, one-second stress duration, one-second rest duration.
Conditions are fixed by equipment manufacturer.
3.2.1. CYCLIC LOAD–DEFORMATION BEHAVIOUR OF KAOLINITE WITH FIBRE AND CEMENT

Deformation of specimens based on actual strain (not taking into account strains
induced during the initial conditioning period) are shown in table 4 and a typical de-
formation–load cycle relationship is shown in figure 6. It can be observed that in rein-
forced specimens the strains are reduced for 0.3% fibre content. This seems to confirm
the earlier observation of increasing fibre content leading to uneven distribution of
fibres. Besides this, kaolinite with 0.3% fibre content also shows the least change in
axial strain from 1.1% to 1.9% (table 4). Cement stabilized specimens show lower
permanent deformation than those containing fibres. However, all the specimens
tested with cement only failed when soaked. For the soaked specimens, the best
strength performance was achieved with 6% cement and 0.3% fibre where at 3,600
cycles permanent strain was less than 1%. Thus it meets the recommendation made by
COOPER et al. [3] for base course and road base mixes. They suggest that resistance to
permanent deformation should not exceed 1% axial strain after 3,600. The results pre-
sented in table 4 also show that the reinforced stabilized specimens were the only
mixes which could undergo soaking tests where the bond strength and friction be-
tween the stabilised clay and fibre were no doubt helping to maintain the integrity of
the specimens.

Table 4
Permanent deformation for kaolinite reinforced with F23 fibre and stabilized with cement
Axial strain
Composites of materials 3,600 cycles 10,000 cycles
Microns % Microns %
Kaolinite 3618 4.5 4199 5.2
Kaolinite + 0.2% F23 849 1.0 1632 2.0
Kaolinite + 0.3% F23 890 1.1 1563 1.9
Kaolinite + 0.4% F23 1024 1.3 2268 2.8
Kaolinite + 4% cement 534 0.7 714 0.9
Kaolinite + 4% cement (soaked specimen) F – F –
Kaolinite + 4% cement + 0.3% F23 2120 2.6 2433 3.0
Kaolinite + 4% cement + 0.3% F23 (soaked specimen) 4127 5.2 4822 6.0
Kaolinite + 6% cement 509 0.6 694 0.9
Kaolinite + 6% cement (soaked specimen) F – F –
Kaolinite + 6% cement + 0.3% F23 1013 1.3 1219 1.5
Kaolinite + 6% cement + 0.3% F23 (soaked specimen) 724 0.9 1125 1.4
F – failed specimen.
 Behaviour of fibre-reinforced and stabilized clayey soils 11

According to DALL’ACQUA [4] the stress level of 100 kPa used to assess the per-
manent deformation of samples immersed in water is too severe if the end use is con-
sidered to be a subbase layer. Nonetheless, the performance of kaolinite reinforced
with 0.3% F23 and stabilized with 6% cement was well above the minimum require-
ments for upper layer of the pavements where the samples were within the 1% limiting
deformation and could withstand the stresses applied. It is also suggested that this
stabilized and reinforced kaolinite can be used for a more demanding structural pave-
ment layer, such as the subbase.

kaolinite+4% cement+0.3% F23_soaked

kaolinite+4% cement+0.3% F23
kaolinite+6% cement+0.3% F23_soaked
kaolinite+6% cement+0.3% F23
kaolinite+6% cement
kaolinite+4% cement

Fig. 6. Relationship between loading cycles and deformation of kaolinite with fibre and cement

3.2.2. CYCLIC LOAD–DEFORMATION BEHAVIOUR OF KAOLINITE

WITH FIBRE AND LIME

The results for kaolinite reinforced with 0.3% fibre and both 4% and 6% lime and
cured for 28 days are shown in table 4, and a typical deformation–load cycle relation-
ship (for 28 days) is depicted in figure 7. The results show that fibres are responsible
for an increase in deformations. This is most probable because of an incomplete
bonding between soil and lime. When lime is added to soil, a clayey component floc-
culates almost immediately. With time, in an environment of high pH, clay minerals
react with calcium to form cementitious products. The cementitious reaction is slow
and after 28 days does not lead to an adequate improvement in strength. This lack of
strength can be observed in figure 7, where specimens with lime are deformed more
than those without.
As with cement-stabilized kaolinite, so with lime-stabilized specimens that fail
when they were submerged in water to assess their durability. In this instance, even
specimens reinforced with fibres failed to survive immersion in water. This was most
12 G.P. DALL’AQUA et al.

probably due to the slow reactivity of lime with clay in terms of strength. However in
practice, early strength gain is important to allow construction to proceed more rap-
idly. Thus, lime was not considered to be a suitable binder for kaolinite in instances,
where early strength is important.
kaolinite+6% lime
kaolinite+4% lime
kaolinite+6% lime+0.3% F23
kaolinite+4% lime+0.3% F23

Fig. 7. Repeated axial load test results for kaolinite with fibre and lime after 28 days

3.3. STRENGTH OF LATERITE

3.3.1. COMPRESSIVE STRENGTH OF LATERITE SUBJECTED

TO MONOTONIC LOADING

Typical stress–strain curves representing the UCS for both plain and reinforced
laterite compacted at OMC and maximum dry density are shown in figure 8. The re-
sults show that there was approximately two-fold increment in strength and strain to
failure when fibre was added. In addition to this, whilst the unreinforced laterite dis-
played a brittle behaviour, reinforced specimen exhibited plastic deformation retaining
about 80% of its strength at 8% strain. According to DALL’ACQUA et al. [4], there was
a significant increase in tensile strength of laterite when fibre was added and that up to
50% of its tensile strength was retained upon soaking. An important result was that
there was no significant reduction in the strength of the specimen containing both fibre
and cement during the soaking tests which indicated the bonds developed between
fibre, cement and soil were substantially water resistant.
Load–deformation curves of laterite, cement-reinforced laterite, cement-stabilized
and fibre-reinforced laterite, unsoaked and soaked specimens are shown in figure 9.
The results show that reinforcement and cement gave the highest strength and that
soaking had little effect on the strength of fibre-reinforced stabilized soil. Laterite
proved to be very durable. This response was similar to that of kaolinite.
 Behaviour of fibre-reinforced and stabilized clayey soils 13

laterite+0.3% F23
laterite

Fig. 8. Compressive stress–deformation curves for plain and reinforced laterite subjected
to unconfined compression

laterite
laterite+6% cement
laterite+6% cement+0.3% F23
laterite+6% cement+0.3% F23_soaked

Fig. 9. Load–deformation curves of laterite, laterite + cement and laterite

with both fibre and cement

3.3.2. BEHAVIOUR OF LATERITE SUBJECTED TO CYCLIC LOADING

In the study of the strength of soaked specimens, only 0.3% fibre and 6% cement
were used. Specimens were prepared as those for kaolinite, described earlier, and also
tested using the NAT. The results of permanent deformation at both 3,600 and 10,000
cycles are shown in table 5 and a typical relationship between deformation and load
cycles is shown in figure 10.
14 G.P. DALL’AQUA et al.

Table 5
Permanent deformation of stabilized and reinforced laterite
Axial strain
Composites of materials 3,600 cycles 10,000 cycles
micron % micron %
Laterite 6210 7.7 9631 11.9
Laterite durability (soaked specimen) C C
Laterite + 0.3% F23 872 1.1 1281 1.6
Laterite + 0.3% F23 (soaked specimen) NS NS
Laterite + 6% cement 875 1.1 1103 1.4
Laterite + 6% cement + 0.3% F23 1100 1.3 1350 1.6
Laterite + 6% cement + 0.3% F23 (soaked specimen) 1362 1.7 1450 1.8
C – collapsed, NS – no strength.

As in the case of kaolinite, the durability of the laterite specimens was assessed by
their curing in the air and then immersing in water. As can be seen in table 5, plain
laterite, reinforced laterite specimens and cement-stabilized specimens were not suit-
able for testing after soaking because the specimens collapsed shortly after soaking.
However, when specimens were stabilized and reinforced, they retained measurable
strength. The results indicate that the change in deformation of reinforced and stabi-
lized laterite is 0.1%, which suggests its little effect in the long term compared to short
term strength. The ability of the specimen to withstand load is mainly due to the fibre
ability to act as a “bridge” for transferring loads to the surrounding soil. With the in-
clusion of cement, shear strength and water resistance of the soil are improved.
laterite
laterite+6% cement+0.3% F23_soaked
laterite+0.3% F23
laterite+6% cement+0.3% F23
laterite+6% cement

Fig. 10. Deformation–load relationship for fibre-reinforced laterite

It is worth noting that fibres hold the soil together even after failure. Thus soils in
a cracked pavement will still retain significant strength. This will prevent the loss of
 Behaviour of fibre-reinforced and stabilized clayey soils 15

soil, particularly by erosion after a dry spell. This may prove particularly useful in the
parts of the world where seasonal climatic effects can lead to the loss of soil used for
pavement construction.

4. CONCLUSIONS

The following conclusions were drawn from this investigation on reinforced and
stabilized kaolinite and laterite subjected to cyclic loading:
• Inclusion of F23 fibres does not have a significant effect on the relationship be-
tween the dry density and moisture content of a soil. However, cement can affect the
dry density.
• In both the soils investigated, fibres resulted in a small increase in both strength
under monotonic loading and strain to failure.
• Reinforced and cement-stabilized specimens showed an improved resistance to
soaking.
• Results showed that cement-stabilized and fibre-reinforced soils may be used in
pavement layers. These findings relate to the soils examined (kaolinite and laterite)
and may apply to other soils.

REFERENCES

[1] British Standards Institution, Methods of tests for soils for civil engineering purposes, BS 1377 Brit-
ish Standard Institution, London, 1990a.
[2] British Standards Institution, Stabilized materials for civil engineering purposes. Methods of test for
cement-stabilized and lime-stabilized materials, British Standard Institution, London, 1990b.
[3] COOPER K.E., BROWN S.F., PRESTON J.N., AKROYD F.M.L., Development of a practical method for
design of hot-mix asphalt, Transportation Research Record 1317, TRB, National Research Council,
Washington, 1991, 42–52.
[4] DALL’ACQUA G.P., Fibre Reinforced Stabilized Soil, PhD Thesis, University of Birmingham, Bir-
mingham, 2000, United Kingdom.
[5] DALL’ACQUA G.P., GHATAORA G.S., FREER-HEWISH R.J., Behaviour of cement stabilised fibre rein-
forced Kaolinite and Laterite, EuroGeo 2000, Bologna, 2000, Italy, Vol. 1, 385–389.
[6] GHATAORA G.S., DALL’ACQUA G.P., FREER-HEWISH R.J., Effect of crimped fibres and lime on the
engineering properties of Kaolinite, 4th International Conference on Ground Improvement – Geo-
systems, Helsinki, 2000, Finland.
[7] GHATAORA G.S., CHAPMAN D.N., SCHAD M., AST W., Influence of Fibres on Properties of Cement
Stabilised Sands and Clay, XIII European Conference on Soil Mechanics and Geotechnical Engi-
neering, Prague, 2003a.
[8] GHATAORA G.S., FREER-HEWISH R.J., DALL’ACQUA G.P., Flexural Strength of Kaolin Clay Rein-
forced with Synthetic Fibres and Stabilised with Lime and Cement, XIII European Conference on Soil
Mechanics and Geotechnical Engineering, Prague, 2003b.
[9] MAHER M.H., HO Y.C., Mechanical properties of Kaolinite fibre soil composite, Journal of Geotech-
nical Engineering, ASCE, 1995, Vol. 120, No. 8, 1381–1393.
16 G.P. DALL’AQUA et al.

[10] PARK T., TAN S.A., Enhanced performance of reinforced soil walls by the inclusion of short fibre,
Geotextiles and Geomembranes, 2005, 23, 348–361.
[11] RANJAN G., VASAN R.M., CHARAN H.D., Probabilistic analysis of randomly distributed fiber-rein-
forced soil, Journal of Geotechnical Engineering, 1996, Vol. 122, No. 6, 419–426.
[12] TANG C., SHI B., GAO W., CHEN F., CAI Y., Strength and mechanical behaviour of short polypropyl-
ene fibre reinforced and cement stabilized clayey soil, 2006, Department of Earth Sciences, Nanjing
University, Nanjing 120093, China.
[13] YETIMOGLU T., SALBAS O., A study on shear strength of sands reinforced with randomely distrib-
uted discrete fibres, Geotextiles and Geomembranes, 2003, 21, 103–110.
[14] WIBISONO G., The effect of water content and velocity on the erodability of a fibre reinforced or
stabilised kaolin, MSc Thesis, The University of Birmingham, Birmingham, 2000, United Kingdom.

View publication stats